Introduction

In 1972, Giedion and colleagues first described CNO as a subacute and chronic symmetric osteomyelitis. In the past 40 years, our understanding of CNO has become more sophisticated because of numerous breakthroughs. The recent advances in imaging technology have led to the enhanced ability to diagnose inflammatory bone lesions. Multiple breakthroughs in immunology have led to a more sophisticated appreciation of the function of the innate immune system. This understanding enabled the characterization of autoinflammatory diseases and, therefore, autoinflammatory bone diseases. There have been many discoveries of the genetic associations of autoinflammatory syndromes. Finally, the use of the tumor necrosis factor α (TNFα) antagonist and bisphosphonates enabled more effective treatments in nonsteroidal antiinflammatory drug (NSAID)–resistant disease. These breakthroughs have enabled the rheumatology community to have a more advanced understanding of CNO.

Nomenclature and Disease Pattern

The terminology for CNO has changed multiple times in the past 40 years. It was first called subacute and chronic symmetric osteomyelitis . However, since that time, it has most commonly been called chronic recurrent multifocal osteomyelitis (CRMO). The term CRMO was coined by Probst and colleagues in 1978 and is characterized as a chronic inflammatory bone disorder that had multifocal bone lesions and had multiple recurrences. However, not all patients have multifocal bone lesions or numerous recurrences. Therefore, the term chronic nonbacterial osteomyelitis has been used as an umbrella term and is inclusive of all the varied presentations of this disease.

There are 3 disease patterns for CNO: a course that resolves within 6 months, a persistent course, and a course characterized by multifocal bone lesions and multiple recurrences. The multifocal recurrent disease pattern is most consistent with CRMO and SAPHO (Synovitis, Acne, Pustulosis, Hyperostosis, and Osteitis) syndrome.

Cause and Pathogenesis of CNO

By definition, the bone lesions seen in CNO are culture negative and have no demonstrable organism on histopathology. Antibiotic therapy should not cause resolution of symptoms. In a small case series, azithromycin was shown to improve radiological and clinical signs and symptoms of CNO. This effect may be mediated through the antiinflammatory properties of azithromycin instead of its antimicrobial properties. Although there have been many proposed pathogens causing CNO, there has been no definitive evidence that microbes trigger CNO. Propionibacterium acnes has been recently proposed as a cause for CNO. This pathogen has been cultured in adults with palmoplantar pustulosis, a common skin manifestation seen in SAPHO and CNO. Rarely was this bacteria cultured when the pustules of patients with palmoplantar pustulosis and SAPHO or CNO were analyzed for P acnes . In one series of adult patients with SAPHO, P acnes was cultured from the bone from 7 of the 15 patients tested. However, in most patients with SAPHO and CNO, cultures of the bone are negative or when positive are thought to be a contaminant.

The chronic inflammation seen with CNO seems to be caused by activation in the innate immune system as typically seen in autoinflammatory diseases. This seemingly unprovoked activation may lead to an imbalance of proinflammatory and antiinflammatory cytokines and, therefore, disruption in immune homeostasis. Currently, the cause and pathophysiology of CNO are not firmly established in nonsyndromic forms of the disease. However, there have been several developments in the possible pathophysiology mechanisms and genetic associations for this disease.

There is evidence that links the interleukin-10 (IL-10) pathway to the development of CNO. Hofmann and colleagues reported that peripheral blood monocytes stimulated with the toll-like receptor 4 agonist lipopolysaccharide (LPS) secreted significantly less IL-10 compared with healthy control monocytes. This outcome occurred independently of IL-10 promoter polymorphisms because an association with CNO and the high IL-10 expressing -1082G>A alleles and the GCC haplotype was found. Because LPS-stimulated CNO monocytes have a decreased production of IL-10, this is the opposite of what would be expected with the high expressing allele, suggesting other mechanisms are involved.

Next, these investigators demonstrated that the decrease in IL-10 secretion from LPS-stimulated CNO monocytes is associated with attenuated extracellular-signal regulated kinase (ERK)1/2 activity. This decrement in ERK1/2 signaling then results in reduced levels of the transcription factor Sp-1, a transcription factor that drives IL-10 gene expression in monocytes. They showed that Sp-1 recruitment to the IL-10.636 Sp-1 element is reduced in LPS-stimulated CNO monocytes. In addition, they found that histone H3 serine-10 phosphorylation (H3S10p), an activating marker, is decreased around the IL-10-636 element of the IL-10 promoter. The attenuation of Sp-1 and reduced H3S10p suggest that epigenetic factors play a role in the decreased gene expression of IL-10 seen in CNO. A unified hypothesis has been proposed to explain these separate pathophysiology mechanisms in CNO. The investigators concluded that impaired mitogen-activated protein kinase signaling, decreased H3S10p, and attenuated Sp-1 recruitment to the IL-10 promoter results in impaired gene expression of IL-10 with subsequent disruption of the proinflammatory and antiinflammatory cytokine balance. This disruption in immune homeostasis might explain part of the clinical presentation of CNO.

There is increasing evidence for the theory that CNO is genetically driven. The two strongest pieces of evidence are that two similar diseases with autoinflammatory bone lesions are genetically driven and animal models with genetic defects have a similar phenotype. Majeed syndrome and DIRA are genetically linked diseases with features that include autoinflammatory bone lesions. Majeed syndrome is caused by mutations in LPIN2 and DIRA with mutations in IL1RN .

There have been many animal models with genetic defects leading to autoinflammatory bone lesions. Currently, there are reports of autoinflammatory bone lesions seen in mice, lemurs, and dogs. Two murine models have developed CNO, the cmo mice and the Lupo mice. Both of these murine models have a mutation in pstpip2 and present with similar clinical features as seen in humans. The cmo mice generally develop a clinically severe presentation. Pstpip2 acts in the cytosol as an F-actin–associated phosphoprotein, interacts with PEST-type protein tyrosine phosphatases, and is involved in cytoskeletal organization. Murine pstpip2 is similar to human PSTPIP1 and PSTPIP2 . PSTPIP1 regulates the NLRP3 inflammasome through binding to pyrin and is the genetic defect seen in the autosomal-dominant autoinflammatory syndrome pyogenic arthritis pyoderma gangrenosum acne syndrome.

There is a possible genetic association locus at chromosome 18q21.3-18q22 and CRMO. This genetic association is not yet linked to the pathophysiological development of CNO. There have been cases of families with multiple affected members with CNO and cases with first- and second-degree relatives with inflammatory bowel disease, psoriasis, and other chronic inflammatory conditions. This finding provides additional evidence of a genetic association. Genetic mutations in PSTPIP 1, PSTPIP2 , CARD15/NOD2 , and IL1RN have been examined in small series and do not seem to be the causative genetic feature in CNO.

Epidemiology

CNO is primarily a disease of childhood. It has many similarities to SAPHO syndrome, which is a disorder primarily seen in adults. The incidence and prevalence of CNO is unknown. Although a diversity of ethnicities and populations throughout the world is affected by CNO, most reports are from Scandinavia, Europe, Australia, and North America. There is a female predominance, and the mean age of disease onset is around 10 years old.

Clinical Presentation

CNO is a diagnosis of exclusion and is established by the clinical presentation, imaging studies, and a culture-negative bone biopsy. Pain with or without swelling at the site of the bony lesion is the typical presenting symptom. Bone lesions tend to cluster around the metaphysis; can occur at atypical locations for bacterial osteomyelitis, such as the clavicle; and when multifocal, often have a symmetric distribution. Seventy-five percent of bone lesions are perimetaphyseal. Appendicular and axial skeletal lesions are also seen. The most common CNO sites are the femur, tibia, pelvis, calcaneus, ankle, vertebrae, and clavicle. CNO is the most common disease cause to affect the medial third of the clavicle in all age groups. A unifocal pattern of disease occurs in 10% to 20% of patients.

CNO is a systemic disease that can affect the skin, joints, gastrointestinal tract, and lungs. Patients with CNO frequently have other coexisting chronic inflammatory diseases. In one study, 20% to 50% of patients had or developed another autoimmune/inflammatory disease. The most frequent associated autoimmune and/or inflammatory diseases included arthritis, psoriasis, inflammatory bowel disease, vasculitis, myositis/fasciitis, and parotitis. Typically, these patients have more bony lesions than patients without a comorbid inflammatory disease.

Patients with CNO tend to have normal, mild, or moderately elevated inflammatory laboratory changes. Markers of inflammation, including erythrocyte sedimentation rate, C-reactive protein, white blood cell count and platelet count, can be moderately increased or completely normal. Inflammatory markers are typically higher in patients with comorbid autoimmune diseases.

Imaging Studies

Radiographs, technetium bone scans, and/or magnetic resonance imaging (MRI) are generally used to detect the lesions and screen for multifocality. Computerized tomography can detect lesions but exposes children to significant levels of radiation, so it is generally not recommended. Typical radiographic findings include a lytic lesion at or around the metaphysis that progresses to sclerosis or hyperostosis. Recently whole-body MRI imaging has been studied as a radiation-free method of imaging. Whole-body MRI imaging is a useful method to detect asymptomatic lesions without radiation exposure. The lesions are best seen on short tau inversion recovery sequences and can be used to identify subclinical spinal involvement, synovitis of adjacent joints, and sacroiliitis. Whole-body MRI tends to poorly detect lesions in the small joints of the hands and feet, ribs, and sternoclavicular and costovertebral joint junctions. There has never been a study to compare bone scintigraphy and whole-body MRI. Whole-body MRI should be considered in working up indeterminate cases of CNO and as an imaging technique used to monitor disease activity and response to therapy.

Histopathology

A bone biopsy is often needed to confirm a diagnosis of CNO. This biopsy is especially necessary in isolated bone lesions because bone malignancies can mimic CNO. In children who present with multifocal disease for many months in duration, the need for a confirmatory bone biopsy is debatable. The presence of clavicular involvement with palmar-plantar pustulosis or psoriasis vulgaris is very strong evidence of CNO. However, caution should be used in the absence of strong supporting evidence of CNO because serious disorders, such as intraosseous lymphoma and other forms of neoplasia, can mimic CNO. A classification criteria and clinical algorithm has been proposed to aid the clinician in the diagnosis of CNO.

CNO is characterized by subacute or chronic inflammation, with a lymphocytic or mixed inflammatory infiltrate, and often marrow fibrosis. By definition, the biopsy is culture negative and has no demonstrable organism on histopathology.

Treatment

NSAIDs are the gold standard initial therapy for CNO. However, there is a discrepancy in the literature regarding the efficacy of NSAIDs in the treatment of CNO. In 2 published studies from Germany, Beck and colleagues demonstrated a complete response to naproxen in 43% of the patients, whereas Girschick and colleagues noted that although 100% of patients with a nonrelapsing course responded to naproxen, only 42% of patients with a relapsing course responded. Indomethacin might be more effective than naproxen but is associated with more side effects. Case studies and small case series have been published addressing the treatment of CNO with various medications, including corticosteroids and disease-modifying antirheumatic drugs (methotrexate [MTX] and sulfasalazine). Because of the small numbers, there is no conclusion about the efficacy of these agents.

TNFα seems to be an important cytokine in CNO. This cytokine plays an important role in the activation of osteoclasts in CNO. There have been several case reports, small retrospective series, and a few larger series on TNFα antagonist treatment in CNO. In a larger series involving 11 patients on TNFα antagonists, 10 of the 11 responded to the therapy, and there was a 46% remission rate. Smaller case series have also showed a favorable response to TNFα antagonists in the treatment of CNO. Eleftheriou and colleagues evaluated 3 pediatric patients treated with TNFα blocking agents with CRMO or SAPHO and all 3 showed a clinical improvement. However, 1 patient stopped therapy prematurely because of an invasive fungal infection. Catalano-Pons and colleagues published the results from a French cohort of 40 pediatric cases of which 2 had used TNFα antagonists, although the treatment response was not thoroughly detailed. Other biologic medications have been used in case reports that include interferon (INF)-α, INF-γ, and anakinra. The case report on anakinra initially showed a response, but the response was not sustained.

Pamidronate has also been recently used for the treatment of CNO. This therapy is hypothesized to treat CNO by inactivating osteoclasts, decreasing pain, and possibly through an antiinflammatory mechanism. In one study, all 9 patients treated responded to pamidronate. The clinical response typically occurred in the first 3 days. Four patients experienced a recurrence in their CNO 12 to 18 months after their first pamidronate course. All 4 patients responded to a repeat course of pamidronate. In another study evaluating the response to pamidronate, 4 of the 5 patients showed clinical improvement. In a third study, 6 of the 7 patients improved with pamidronate. However, synovial joint disease was unresponsive to the therapy. Pamidronate therapy may be particular beneficial in spinal lesions and vertebral fractures by improving vertebral shape and decreasing the kyphotic angle. Hospach and colleagues reported on patients with axial disease and found that all 7 had an improvement in spinal lesions after pamidronate therapy. Urinary N-telopeptide/urine creatinine (uNTX/uCr), a marker of collagen-I breakdown, has been proposed as a marker for disease flare after bisphosphonate therapy. This marker has been used to monitor disease with accelerated bone turnover. When used in cases with CNO, no clinically evident relapses occurred while uNTX/uCr was suppressed.

Concerns have been raised about using bisphosphonates in the pediatric population because long-term safety data are limited. The most common adverse events are minor flulike symptoms for a day after the infusion. Currently, osteonecrosis of the jaw is possible but is an unreported side effect of pamidronate use in children with CNO. This adverse event primarily occurs in elderly patients with myeloma. Because of this potential side effect, it is recommended to have pediatric patients have a dental screening and wisdom teeth extraction before pamidronate therapy whenever possible and postpone elective dental procedures for at least 6 months following therapy.

No study has compared the efficacy of bisphosphonates with TNFα antagonists. There have been 2 larger observational studies that treated patients with either bisphosphonates or TNFα antagonists (3 and 2 in one cohort and 3 and 1 in the other cohort), but no data were included on the efficacy of these therapies ( Fig. 1 ).

Only gold members can continue reading. Log In or Register to continue

Share this:

• Click to share on Twitter (Opens in new window)
• Click to share on Facebook (Opens in new window)

Related

Related posts:

Treatment of Juvenile Idiopathic Arthritis in the Biologic Age Update on Juvenile Spondyloarthritis Update in Pediatric Inflammatory Bowel Disease Biomarkers and Updates on Pediatrics Lupus Nephritis Developments in the Classification and Treatment of the Juvenile Idiopathic Inflammatory Myopathies New Developments in Juvenile Systemic and Localized Scleroderma

Stay updated, free articles. Join our Telegram channel

Join